Random Forest Model in the Diagnosis of Dementia Patients with Normal Mini-Mental State Examination Scores.

Jie Wang,Jianyong Wang,Dan Lei,Shanshan Chu,Caiyan Liu,Xinying Huang,Chenhui Mao,Jie Li,Zhuo Wang,Ning Liu,Jing Gao,Liling Dong

doi:10.3390/jpm12010037

Abstract

Background: Mini-Mental State Examination (MMSE) is the most widely used tool in cognitive screening. Some individuals with normal MMSE scores have extensive cognitive impairment. Systematic neuropsychological assessment should be performed in these patients. This study aimed to optimize the systematic neuropsychological test battery (NTB) by machine learning and develop new classification models for distinguishing mild cognitive impairment (MCI) and dementia among individuals with MMSE ≥ 26. Methods: 375 participants with MMSE ≥ 26 were assigned a diagnosis of cognitively unimpaired (CU) (n = 67), MCI (n = 174), or dementia (n = 134). We compared the performance of five machine learning algorithms, including logistic regression, decision tree, SVM, XGBoost, and random forest (RF), in identifying MCI and dementia. Results: RF performed best in identifying MCI and dementia. Six neuropsychological subtests with high-importance features were selected to form a simplified NTB, and the test time was cut in half. The AUC of the RF model was 0.89 for distinguishing MCI from CU, and 0.84 for distinguishing dementia from nondementia. Conclusions: This simplified cognitive assessment model can be useful for the diagnosis of MCI and dementia in patients with normal MMSE. It not only optimizes the content of cognitive evaluation, but also improves diagnosis and reduces missed diagnosis.

Highlights

The prevalence of dementia is rising with the aging of the population, affecting the quality of life and increasing the burden on society and the family [1]
The dementia group was significantly older than the mild cognitive impairment (MCI) group, and years of education were significantly higher in the cognitively unimpaired (CU) than in the subjects with MCI and dementia
The present study found that 35.7 percent of subjects with Mini-Mental State Examination (MMSE) scores ≥ 26 had evidence of dementia

Summary

Introduction

The prevalence of dementia is rising with the aging of the population, affecting the quality of life and increasing the burden on society and the family [1]. Mild cognitive impairment (MCI) is considered a transitional stage between normal aging and dementia, with a higher risk of developing dementia. The most widely used screening tool for dementia is the Mini-Mental State Examination (MMSE) [4], a 30-point instrument that assesses several domains including orientation, attention, language, memory, and executive function. Pooled estimates of 15 studies showed a sensitivity of 0.89 and specificity of 0.89 at a cut point of or less or or less [6]. The sensitivity (0.20–0.93) and specificity (0.48–0.93) to detect MCI vary significantly in different studies, meaning less consistent estimates for test accuracy [6]. This study aimed to optimize the systematic neuropsychological test battery (NTB) by machine learning and develop new classification models for distinguishing mild cognitive impairment (MCI) and dementia among individuals with MMSE ≥

Objectives

Methods

Results

Discussion

Conclusion